microsoft_clarity-automation
by ComposioHQmicrosoft_clarity-automation helps agents use Microsoft Clarity through Composio Rube MCP for session recordings, heatmaps, and behavior analytics. It emphasizes RUBE_SEARCH_TOOLS first, active Clarity connection setup, and schema-aware usage.
This skill scores 68/100, which means it is acceptable for directory listing but should be presented as a lightweight MCP workflow guide rather than a fully self-contained automation package. Directory users get enough evidence to understand when to use it—Microsoft Clarity automation through Composio/Rube—and how to start safely, but should expect to rely on live tool discovery for exact schemas and execution details.
- Valid skill frontmatter clearly names the target automation domain and declares the required Rube MCP dependency.
- Prerequisites and setup steps explain how to connect Rube MCP, manage the Microsoft Clarity connection, and confirm ACTIVE status before use.
- The skill explicitly instructs agents to call RUBE_SEARCH_TOOLS first for current Microsoft Clarity schemas, improving triggerability against changing Composio tool definitions.
- No support files, examples, scripts, or reference materials are included beyond SKILL.md, so adoption depends heavily on live Rube MCP tool discovery.
- The excerpt shows high-level workflow intent but limited concrete Microsoft Clarity task examples or expected inputs/outputs, which may leave agents guessing after schemas are discovered.
Overview of microsoft_clarity-automation skill
What microsoft_clarity-automation does
The microsoft_clarity-automation skill helps an AI agent operate Microsoft Clarity through Composio’s Rube MCP toolkit. It is designed for workflows around session recordings, heatmaps, project analytics, user behavior review, and Clarity data retrieval without manually guessing tool names or schemas.
Its most important instruction is operational: the agent should call RUBE_SEARCH_TOOLS first, because Microsoft Clarity tool schemas can change. That makes this skill more useful than a static prompt when you need a live, schema-aware workflow.
Best-fit users and jobs
This skill is a good fit for growth, product, UX, and analytics teams that already use Microsoft Clarity and want an agent to help inspect behavioral data. Typical jobs include finding relevant recordings, checking heatmap availability, summarizing user friction, or preparing an analysis workflow before deeper manual review.
It is especially relevant for users building Workflow Automation around customer behavior research, conversion analysis, onboarding reviews, or website issue triage.
Key adoption requirements
To use the microsoft_clarity-automation skill, your AI client must support MCP and must have Rube MCP configured. You also need an active Microsoft Clarity connection through RUBE_MANAGE_CONNECTIONS using the microsoft_clarity toolkit.
The upstream repository contains a single primary file, SKILL.md, with no bundled scripts, references, or helper resources. This keeps installation simple, but it also means your prompt must supply the business goal, project context, date range, and analysis criteria.
How to Use microsoft_clarity-automation skill
microsoft_clarity-automation install and setup path
Install from the Composio skill collection:
npx skills add ComposioHQ/awesome-claude-skills --skill microsoft_clarity-automation
Then open the installed SKILL.md at:
composio-skills/microsoft_clarity-automation/SKILL.md
Configure Rube MCP in your client by adding:
https://rube.app/mcp
Before asking for Clarity work, verify that RUBE_SEARCH_TOOLS is available. Then call RUBE_MANAGE_CONNECTIONS with toolkit microsoft_clarity. If the connection is not ACTIVE, complete the returned authorization flow and confirm active status before running analysis tasks.
Inputs the skill needs
A weak request like “analyze Clarity” leaves too much ambiguity. A strong microsoft_clarity-automation usage prompt should include:
- the Microsoft Clarity project or site you care about
- the business question, such as “why users abandon pricing”
- the date range or comparison period
- the behavior signals to inspect, such as rage clicks, scroll depth, dead clicks, session recordings, or heatmaps
- output format, such as a prioritized issue list, UX research notes, or experiment ideas
- any privacy, compliance, or reporting constraints
Example prompt:
“Use microsoft_clarity-automation to inspect Microsoft Clarity data for the marketing site. First discover current Rube tools and schemas. Focus on the last 14 days, especially pricing and signup pages. Look for heatmap patterns, confusing clicks, and session recording evidence of form friction. Return a prioritized table with issue, evidence, affected page, confidence, and recommended next action.”
Practical workflow that reduces failures
Start every run with tool discovery:
RUBE_SEARCH_TOOLS: queries=[{"use_case":"session recordings, heatmaps, and user behavior analytics","known_fields":""}]
Use the returned tool slugs and schemas rather than inventing parameters. Next, confirm the Microsoft Clarity connection is active. Then run the smallest useful query first, such as one project, one page group, or one date range. Expand only after the initial result proves the schema and data access are correct.
For best results, ask the agent to separate “observed evidence” from “interpretation.” Clarity data can show behavior patterns, but it does not always prove user intent.
Repository files to read first
For this skill, SKILL.md is the file that matters. It includes prerequisites, setup guidance, tool discovery instructions, and core workflow framing. There is no README.md, metadata.json, rules/, resources/, references/, or scripts/ folder in the provided structure, so do not expect packaged examples beyond the skill instructions.
microsoft_clarity-automation skill FAQ
Is microsoft_clarity-automation for beginners?
Yes, if your MCP client is already set up and you can complete the Microsoft Clarity authorization flow. The skill reduces the need to know Composio tool names in advance because it instructs the agent to search available tools first.
It is less beginner-friendly if you have never configured MCP tools, because the main blocker is not the skill text; it is connecting Rube MCP and activating the Microsoft Clarity toolkit.
How is it better than an ordinary prompt?
An ordinary prompt may hallucinate Microsoft Clarity API calls or outdated fields. The microsoft_clarity-automation guide pattern is stronger because it tells the agent to discover current tool schemas through RUBE_SEARCH_TOOLS before execution.
That matters when automating analytics work: wrong parameters can produce empty results, misleading summaries, or failed tool calls.
When should I not use this skill?
Do not use it if you only need a human-readable explanation of what Microsoft Clarity is, or if you do not have access to the target Clarity project. It also is not a replacement for product analytics instrumentation, A/B testing, or compliance review.
Avoid using it for broad, vague requests such as “tell me what users think.” Microsoft Clarity behavior data is strongest when tied to specific pages, funnels, time windows, and observable events.
Does it fit Workflow Automation stacks?
Yes. microsoft_clarity-automation for Workflow Automation is most useful when combined with repeatable analysis routines: weekly UX review, launch monitoring, conversion issue triage, or support-ticket investigation. Pair it with clear reporting templates so the output can feed product, design, or growth workflows.
How to Improve microsoft_clarity-automation skill
Improve microsoft_clarity-automation prompts
The biggest quality gain comes from making the goal measurable. Instead of asking for a broad analysis, define the decision you need to make.
Better input:
“Review Clarity behavior for the checkout funnel after the new form release. Compare the last 7 days with the previous 7 days if tools support it. Prioritize issues that appear in recordings or heatmaps and could block purchase completion.”
This gives the agent a workflow, a comparison frame, and a ranking criterion.
Common failure modes to prevent
The most common failure is skipping tool discovery and guessing schemas. Prevent this by explicitly saying: “Call RUBE_SEARCH_TOOLS first and use only returned tool schemas.”
Another failure is over-interpreting behavior. Ask the agent to label confidence levels and cite the Clarity signal behind each finding. If the data is unavailable, the agent should say what is missing rather than invent a conclusion.
Iterate after the first output
Treat the first result as a discovery pass. Follow up with narrower prompts such as:
- “Show only high-confidence issues affecting signup.”
- “Group findings by page template.”
- “Turn the top three issues into experiment hypotheses.”
- “List what additional Clarity data would strengthen or weaken these conclusions.”
This makes the skill more useful for action, not just summary.
Add local team context
Because the repository does not include custom business rules, add your own context in the prompt: key pages, funnel definitions, known releases, excluded traffic, device focus, and reporting format. For recurring use, save a team-specific prompt wrapper that always includes project name, date range, success metric, and required evidence format.